Simple and feasible detection of hepatitis a virus using reverse transcription multienzyme isothermal rapid amplification and lateral flow dipsticks without standard PCR laboratory

Abstract

Hepatitis A virus (HAV) is mainly transmitted via contaminated food and water. HAV infection is a major global public health problem. Thus, developing a simple, rapid detection method is crucial for containing HAV epidemics, particularly in developing regions with limited laboratory resources. This study established a feasible HAV detection solution by combining reverse transcription multienzyme isothermal rapid amplification (RT-MIRA) and lateral flow dipstick (LFD) strips. Primers targeting the conserved 5’UTR sequence of HAV were used in the RT-MIRA-LFD assay. RNA extraction was enhanced by obtaining RNA directly from the centrifuged supernatant. Our study found that MIRA amplification could be finished in 12 min at 37 °C and naked-eye observation of the LFD strips in 10 min. The detection sensitivity of this method reached 1 copy/μl. RT-MIRA-LFD was compared to conventional RT-PCR using 35 human blood samples. The accuracy of the RT-MIRA-LFD method was 100%. The convenience, sensitivity, and rapidness of this detection method could provide a considerable advantage for diagnosing and controlling HAV infection, especially in regions with limited medical resources.

Keywords:

• Hepatitis A virus
• reverse transcription multienzyme isothermal rapid amplification
• lateral flow strip
• nucleic acid detection
• rapid diagnosis

Introduction

Hepatitis A virus (HAV), classified as hepatovirus, is a small, unenveloped symmetrical RNA virus. It belongs to the micro ribonucleic acid viruses and is the cause of infectious or epidemic hepatitis [1]. HAV spreads through the fecal-oral route and is commonly contracted by ingesting food or water contaminated with the faeces of infected patients and vegetables fertilised with faeces and uncooked shellfish [2–4]. In developing regions, the source pollution of primary and processed agricultural products is still a serious problem. Although, the infection of HAV is more common in developing countries, an increased number of HAV cases has been reported in developed countries [5]. In developed cities, food may also be contaminated by unkilled HAV virus [6]. A 2019 HAV-mediated outbreak in the United States was linked to imported fresh blackberries [7]. HAV infections cause nausea, vomiting, abdominal pain, jaundice, fever, and diarrhoea [8]. Approximately 10 to 15% of infected patients will have recurring symptoms within six months of the initial infection. In severe cases, acute liver failure may be life-threatening [9]. Moreover, HAV infections during community-based, person-to-person outbreaks cannot be ignored [10].

HAV infection is a global public health problem. Around the world, 1.5 million people are infected with HAV each year, and the number of undiagnosed infections may reach tens of millions [11]. In China, hepatitis A outbreaks are a public health crisis, causing great panic and a huge shock to the economy. In 1988, 1.5 million people in Shanghai were infected by HAV transmitted by Scapharca subcrenata, with more than 300,000 presenting with symptoms [6]. In that year, the cost of free treatment for hepatitis patients far exceeded the health expenditure of Shanghai [12]. Because Hepatitis A often spreads through contaminated food and water, it can infect many people in a short time (a cluster outbreak) [13]. Therefore, a rapid, accurate, and economical diagnostic test for HAV is urgently needed to control hepatitis A epidemics, especially in resource-limited settings [14].

The current gold standard for diagnosing HAV infections is an enzyme-linked immunosorbent assay (ELISA) based on serum anti-HAV IgM levels, which is less effective in detecting recent infections because of a stringent window for initial tests [15]. During an HAV outbreak, when a rapid and precise diagnosis is essential, molecular assays could be introduced to supplement or confirm serology results. Presently, reverse transcription polymerase chain reaction (RT-PCR) is a widely used molecular detection method in the clinic [16]. However, it requires high-precision and expensive devices. Due to limited laboratory facilities and professionals, it cannot be popularised in resource-limited regions.

Several isothermal technologies have been developed for rapid, feasible nucleic detection. Multienzyme isothermal recombinase amplification (MIRA) is a novel nucleic acid amplification method based on recombinase polymerase amplification (RPA) [17]. This technique utilises four core proteins: recombinase-Rec A, DNA helicase-GP41, single-stranded binding (SSB) protein, and DNA pol I [18]. During amplification, helicase-GP41 and SSB form the D-Loop and start the reaction. Recombinase and DNA pol I then allow DNA extension rapidly at isothermal conditions. For RNA template amplification, specific primers can be used to directly synthesise cDNA using reverse transcriptase [19,20]. Simultaneously, the newly generated cDNA is used as a template for MIRA [21]. The whole reaction can be performed at 25 to 42 °C within 5 to 30 min without requiring complex instruments, providing a major advantage over conventional PCR. The lateral flow dipstick (LFD) can produce visually observable results in a relatively short time by hybridising with labelled colloidal gold [22]. Moreover, it provides a convenient, sensitive, and economical assay when combined with RT-MIRA.

To solve the current problems associated with HAV detection and popularise early diagnosis of HAV infection in resource-limited regions, our study integrated RT-MIRA and LFD dipstick technologies to generate a convenient, sensitive, and economical method of HAV diagnosis.

Method

Primer design

The HAV 5′-untranslated region (5′-UTR) is 734 bp, accounts for about 10% of the whole genome, and is the most conserved sequence [16,23]. Therefore, we designed primers targeting 5′-UTR region using Premier 5.0 software, according to the guidelines of AMP-Future Biotech Co. Ltd. (Weifang, China) (Figure 1). The primer lengths were about 25 to 35 bp, and the amplicon was approximately 150 to 300 bp (Table 1). Specificity analysis was performed using BLAST (http://www.ncbi.nlm.nih.gov/blast/Blast.cgi).

Figure 1. The primer targeting regions.

Table 1. The primers used for MIRA.

Primer	Sequence (5’ to 3’)	Position in HAV sequence*	Product length (bp)
F-a1	AGGGGACTTGATACCTCACCGCCGTTTG	53–80	180
R-a1	TGGAAGAAAGAAGACAGAAAGCGTGAAAAATGAG	199–232
F-a2	CACTCATTTTTCACGCTTTCTGTCTTCTTTC	197–227	269
R-a2	CCGCCGCTGTTACCCTATCCAAGGCATCT	437–465
F-a1 labelled	[5′-FAM]-AGGGGACTTGATACCTCACCGCCGTTTG	53–80	180
R-a1 labelled	[5′-biotin]-TGGAAGAAAGAAGACAGAAAGCGTGAAAAATGAG	199–227

*NCBI Reference Sequence: NC_001489.1.

HAV 5’UTR template preparation

The HAV 5′-UTR was cloned into pCDNA 3.1(+) by TSINGKE Biotech Co. Ltd. (Nanjing, China). After plasmid extraction, the DNA concentration was measured using a NanoDrop^TM One UV spectrophotometer (ThermoFisher Scientific, Waltham, MA), and the copy number was calculated according to Kozak et al. [24].

MIRA

MIRA was performed using the DNA isothermal rapid amplification kit (#WLB8201KIT, AMP-Future Biotech Co. Ltd), according to the manufacturer’s instructions. Two primer pairs (First pair: F-a1 and R-a1; Second pair: F-a1 ad R-a2) were designed to screen the appropriate (Table 1). The reaction was performed at 37 °C for 30 min. The generated products (5 μl) were purified with phenol and separated on a 2% agarose gel at 120 V for 1 h. After electrophoresis, the products were visualised using an automatic digital gel image analysis system (Tanon, Shanghai, China). Water was used as a negative control, and the template DNA provided in the kit served as the positive control.

MIRA-LFD

Forward and reverse primers (name: F-a1 labelled and R-a1 labelled) were labelled at the 5′-end with FAM or biotin, respectively (Table 1). MIRA-LFD was performed using an amplification kit (Colloidal gold test strip type) (#WLB8203KIT, AMP-Future Biotech Co. Ltd), according to the manufacturer’s instructions. The reactions were performed at 37 °C for 12 min, and the products were diluted 1:5000 with H₂O. Aliquots of the diluted products (100 μl) were applied to LFDs (#JY0201, Baoying Tonghui Biotechnology Co. Ltd., Beijing, China), and the LFDs were visually observed after 10 min.

Sensitivity and specificity analysis

To determine the sensitivity of the MIRA reaction, HAV 5′-UTR plasmid DNA was serially diluted with ddH₂O in 10-fold increments from 10 pg/μl to 10 fg/μl, and the MIRA products were separated by electrophoresis using a 2% agarose gel. In parallel, the dilutions were subjected to MIRA-LFD and qPCR. qPCR was performed with TaKaRa TB Green Premix Ex Taq II (Tli RNaseH Plus, TaKaRa, Dalian, China) using the Applied Biosystems 7500 Real-Time PCR system (Thermo Fisher Scientific). The qPCR primers were previously reported [25]. The reaction conditions were as follows: 95 °C for 30 s, 40 cycles at 95 °C for 5 s and 60 °C for 34 s. The templates consisted of HAV 5′-UTR plasmid DNA diluted from 10 pg/μl to 10⁻² fg/μl. All the amplification reactions were performed in triplicate.

For specificity analysis, the cross-reactivity of the MIRA assay was determined using common infectious viruses, including hepatitis B virus (HBV), hepatitis C virus (HCV), hepatitis E virus (HEV), human immunodeficiency virus-1 (HIV-1), herpes simplex virus (HSV-1), Japanese encephalitis virus (JEV) and rotavirus. All the templates used in specificity analysis were 1 ng/μl plasmid DNA contains the target virus sequence. We also evaluated the specificity of the MIRA-LFD reaction using the human genome as the template, as previously reported [17,26].

RT-MIRA

For better quantitative analysis and experimental safety, we used HAV live-attenuated vaccine strain H2 instead of live hepatitis A virus [27]. Due to a mutation in the 5′-UTR of strain H2 [28] resulting in virus attenuation and cell adaptation, we adjusted the reverse primer sequences accordingly (primer name: F-a1A and R-a1A)(Table 2). RNA was extracted from strain H2 using the rapid nucleic acid release agent (RNA type) (#WLR8203, AMP-Future Biotech Co. Ltd., Weifang, China). The extracted RNA was purified using the RNAclean Kit (DP412, Tiangen, China) to remove the vaccine excipients, which can inhibit the isothermal amplification.

Table 2. The primers used for RT-MIRA of HAV vaccine strain H2.

Primer	Sequence (5’ to 3’)	Position in H2 stain sequence	Product length (bp)
F-a1A	AGGGGACTTGATACCTCACCGCCGTTTG	53–80	175
R-a1A	TGGAAGAAAGAAGACAGAAAGCGTGAAAATGAG	195–227
F-a1A labelled	[5′-FAM]-AGGGGACTTGATACCTCACCGCCGTTTG	53–80	175
R-a1A labelled	[5′-biotin]-TGGAAGAAAGAAGACAGAAAGCGTGAAAATGAG	195–227

RT-MIRA was performed using the RNA isothermal rapid amplification kit (#WLB8207KIT, AMP-Future Biotech Co. Ltd). Amplification and analysis were performed as described for MIRA, using water as a negative control and plasmid DNA as a positive control.

RT-MIRA-LFD

Different from the primers used in RT-MIRA reaction, the forward and reverse primers were labelled, respectively (primer name: F-a1A labelled and R-a1A labelled) (Table 2). RNA template was also extracted from strain H2 using the rapid nucleic acid release agent. RT-MIRA-LFD was performed using an amplification kit (Colloidal gold test strip type) (#WLB8209KIT, AMP-Future Biotech Co. Ltd). Amplification and detection using LFDs were as described for MIRA-LFD.

DNA sequencing

The MIRA products were sequenced by Sanger DNA sequencing using the ABI3730XL (Taihe Biotechnology Co. Ltd., Beijing, China). Primer F-a1 was used for sequencing (Table 2).

Clinical sample analysis

A total of 35 peripheral blood samples were collected at local hospitals, of which 19 were identified as HAV-positive by RT-PCR [29]. RNA was extracted from 0.5 ml peripheral blood using the rapid nucleic acid release agent (#WLR8203, AMP-Future Biotech Co. Ltd., Weifang, China). The extracted RNA was then amplified using the RT-MIRA-LED kit and detected with LFDs (lateral flow nucleic acid test notes).

Results

Selection of primers

We validated two primer pairs designed to recognise the HAV 5′-UTR (Table 1, Figure 2). To optimal primer pair was determined by performing MIRA on 10-fold serial dilutions of HAV 5′-UTR plasmid DNA (Figure 3). The results showed that although both the a1 and a2 primers could detect templates at concentrations ranging from 10⁴ copies/μl to 10⁷ copies/μl, the a1 primers appeared to be more efficient, producing more intense bands. Therefore, the a1 primers were used for the subsequent experiments. The sequence of the MIRA product generated using the a1 primers is shown in Figure 4. No other pathogens tested could be detected by MIRA using the a1 primers (Figure 5).

Figure 2. MIRA products separated on a 2% agarose gel. L: 50-bp DNA Ladder; 1: positive control (∼250 bp); 2: primer a1 (∼180 bp); 3: primer a2 (∼269 bp); 4: primer a1, negative control; 5: primer a2, negative control.

Figure 3. The MIRA products for the serial dilution of HAV plasmid DNA separated on a 2% agarose gel. L: 50-bp DNA Ladder; 1: positive control (∼250 bp); 2: 10⁷ copies/μl template with a1 primers; 3: 10⁶ copies/μl template with a1 primers; 4: 10⁵ copies/μl template with a1 primers; 5: 10⁴ copies/μl template with a1 primers; 6: 10⁷ copies/μl template with a2 primers; 7: 10⁶ copies/μl template with a2 primers; 8: 10⁵ copies/μl template with a2 primers; 9: 10⁴ copies/μl template with a2 primers.

Figure 4. The Sanger sequence of the HAV plasmid DNA product generated using primer a1.

Figure 5. Cross-reactivity testing against multiple pathogens using MIRA. L: 50-bp DNA Ladder; 1: positive control (∼250 bp); 2: HAV; 3: HBV; 4: HCV; 5: HEV; 6: HIV-1; 7: HSV-1; 8: JEV; 9: rotavirus 10: human genome.

MIRA-LFD

We performed 10-fold serial dilutions of HAV5’-UTR plasmid DNA (10⁵ copies to 0.1 copy/μL) and analysed the samples using MIRA-LFD. We found that the test strips could detect plasmid DNA at concentrations as low as 0.1 copy/μL (Figure 6). In contrast, the water and human genome template samples yielded negative results. In parallel, we analysed 10-fold serial dilutions of plasmid from 10⁷ copies to 10 copy/μL using qPCR, which had a detection limit of 10² copies (Figure 7). These results indicated that MIRA-LFD was more sensitive than qPCR for detecting HAV.

Figure 6. Serial dilution of HAV plasmid DNA detected by MIRA-LFD. 1: water template; 2: human genome template; 3: 10⁶ copies/μl template; 4: 10⁵ copies/μl template; 5: 10⁴ copies/μl template; 6: 10³ copies/μl template; 7: 10² copies/μl template; 8: 10 copies fg/μl template; 9: 1 copy/μl template; 10: 0.1 copy/μl template.

Figure 7. Serial dilution of HAV plasmid DNA detected by qPCR. 1: 10⁷ copies/μl template; 2: 10⁶ copies/μl template; 3: 10⁵ copies/μl template; 4: 10⁴ copies/μl template; 5: 10³ copies/μl template; 6: 10² copies/μl template; 7: 10 copies/μl template; 8: water template.

RT-MIRA

RT-MIRA was performed on 10-fold serial dilutions of strain H2 RNA (10⁶ copies/μl to 10² copies/μl) (Figure 8). All templates could be detected by this method. The sequence of the RT-MIRA product is shown in Figure 9.

Figure 8. RT-MIRA products for the serial dilution of HAV vaccine strain H2 template separated on a 2% agarose gel. L: 50-bp DNA Ladder; 1: plasmid DNA template, positive control; 2: water template, negative control; 3: 10⁶ copies/μl template; 4: 10⁵ copies/μl template; 5: 10⁴ copies/μl template; 6: 10³ copies/μl template; 7: 10² copies/μl template

Figure 9. The Sanger sequence of the HAV vaccine strain H2 product generated by RT-MIRA.

RT-MIRA-LFD

Similar to the MIRA-LFD results, the concentrations of strain H2 from 10⁵ copies/μl down to 1 copy/μl were detected using RT-MIRA-LFD. In contrast, nothing was detected in the human genome template sample (Figure 10). The limit of detection is 1.56 ± 0.28 copies/μl (95% CI = 0.7916-2.3284).

Figure 10. Serial dilution of HAV vaccine strain H2 template detected by RT-MIRA-LFD. 1: water template; 2: human genome template; 3: 10⁵ copies/μl template; 4: 10⁴ copies/μl template; 5: 10³ copies/μl template; 6: 10² copies/μl template; 7: 10 copies/μl template; 8: 1 copy/μl template; 9: 0.1 copy/μl template.

Clinical sample analysis

The RT-MIRA-LFD assay was validated using 19 positive and 16 negative clinical samples. The results obtained by RT-MIRA-LFD were consistent with the results from traditional RT-PCR (Table 3), indicating that RT-MIRA-LFD has great potential as a clinical application.

Table 3. HAV detection in clinical samples.

		Real-time PCR			Accuracy rate
		positive	negative	total
MIRA-LFD	positive	19	0	19	100%
	negative	0	16	16
	total	19	0	35

Discussion

We established a new field-adapted HAV detection method by combining RT-MIRA and LFD. The scheme of our method is summarised in Figure 11. HAV RNA is extracted from blood samples using a rapid RNA release agent. The RNA can be obtained directly from the supernatant after moderate vortexing and centrifugation. The entire RT-MIRA reaction took only 12 min at 37 °C. The product was then diluted and applied to the test strip, with positive results visually observed after 10 min. This method could effectively avoid the disadvantages of existing detection technology and does not require complex equipment or operation. Thus, it is suitable for promotion and popularisation in resource-limited regions. Moreover, after qualification according to Kim’s method[30], its detection sensitivity even reached 1 copy/μl in clinical samples, indicating this assay could play a significant role in preventing and controlling the spread of HAV infection.

Figure 11. Scheme summarising of the complete testing process.

With the rapid development of molecular biomedical technologies in recent years, several isothermal amplification solutions for both DNA and RNA have been advanced. Nuclear acid sequence-based amplification (NASBA) and simultaneous amplification and testing (SAT) are two current RNA isothermal amplification technologies. Both techniques involve a complex process of continuous transcription and reverse transcription between cDNA and RNA to achieve nucleic acid amplification. NASBA relies heavily on manual handling at 42 °C [31], while SAT requires repeated addition of reverse transcriptase and T7 RNA polymerase [32]. Thus, both techniques require specialised experts and are unsuitable for resource-limited regions.

For HAV detection, real-time reverse transcription loop-mediated isothermal amplification (RT-LAMP) can be finished in 50 min at 62.5 °C [33]. Multiple cross displacement amplification (MCDA) [34–36], cross-priming amplification (CPA) [37], and RT-LAMP [38–41] have been proved to detect various pathogens at about 60 °C within one hour. In addition, of all these methods, more than four primers need to be designed which is complicated and the verification steps of amplicons are also complicated. RPA technique, which also depends on multiple enzymes, is the foundation of MIRA [18,42]. The reaction can be performed at 37 ∼ 42 °C for 20 ∼ 40 min and only needs one pair of primers. However, RPA is not ideal for distinguishing mutations or identifying mutations based on nested-RPA [43], and the cost of RPA assay could reach 12 dollars/reaction. Some PCR-based techniques just like real time PCR, rapid single-tube protocol using nested real-time PCR and reverse transcription-droplet digital PCR method for quantitative detection have been reported [25,41,44–46]. Unfortunately, these methods cannot be widely popularised in remote, less-developed areas due to the high price of the machines.

Our RT-MIRA method only requires 12 min at 37 °C, which is important for the rapid detection of HAV to control the spread of the virus. RT-MIRA is a highly sensitive and selective isothermal amplification technique with minimal sample preparation and is capable of amplifying as low as 1 to 10 DNA target copies in less than 20 min [47]. Additionally, it does not require the strict laboratory conditions of earlier methods [18]. At room temperature, a modified reverse transcriptase synthesises cDNA strands using specifically designed primers and template RNA. Recombinase, SSB protein, and DNA polymerase in the reaction system use the newly synthesised cDNA strands as templates for rapid nucleic acid amplification [17]. Furthermore, the reverse transcription and amplification steps are carried out simultaneously [21], with the product cDNAs amplified directly, greatly simplifying the process, and speeding up the analysis. After amplification, an LFD strip can be used to detect the products using labelled primers.

The whole detection process can be completed within 30 min compared to three hours for ELISA, four hours for RT-PCR and one hour for RT-LAMP. The cost of MIRA reaction is about 6.5 dollars/sample, which is roughly half that of the RPA reaction. Moreover, MIRA reaction does not require an expensive PCR instruments, and only requires constant temperature reaction at 37 °C for 12 min. And if the time cost and the difficulty of primer design was considered, the total cost of each reaction of has no obvious difference compared with RT-LAMP [48]. Therefore, this technique is a feasible approach for detecting potential HAV infections via human blood samples, which could improve the diagnosis and control of HAV [22].

The current study only tested our new RT-MIRA assay using blood samples; however, stool and food samples also play vital roles in HAV diagnosis and disease control [49]. Thus, further studies are essential for assessing the feasibility of using this RT-MIRA-LFD on other biological sample types.

In summary, the combination of RT-MIRA and LFD is a convenient, sensitive, and economical approach that does not require special instrumentation. The entire RT-MIRA-LFD process only takes 12 min at 37 °C, and the results can be visualised ten minutes after applying the reaction products to the test strip. Therefore, this method has great potential for use in resource-limited regions, especially in high HAV endemic areas.

Author contributions

All the authors involved played an important role and contributed to the article. All authors approved the submitted version. YQP and JY contributed to the research design. MLS and YZ analysed the data. MLS provided clinical samples and conducted the molecular biology experiments. JY performed the experiments and provided study supervision. YQP and MLS wrote the manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data sets used and/or analysed during this study are available from the corresponding author on reasonable request.

Additional information

Funding

This study was supported by 345 Talent Project of Shengjing Hospital of China Medical University.